[Bug target/93588] Vectorized load followed by FMA pessimizes on Haswell from version 8.1

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93588

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |x86_64-linux-gnu
             Status|UNCONFIRMED                 |WAITING
   Last reconfirmed|                            |2020-02-05
     Ever confirmed|0                           |1

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
>The intrinsics are supposed to map to the corresponding assembly instructions, no?
NO, it is an interface to what the assembly instructions do; not always 1-1
mapping.  Have you benchmarked both versions?

Also what happens if you use -march=native on your machines?  Do the benchmark
for that version is the best?